Alle Publikationen
2019
-
(2019): Künstliche Intelligenz - Wann übernehmen die Maschinen?. 2. Aufl.. Berlin, Heidelberg: Springer Berlin Heidelberg
Abstract: Einführung: Was ist KI? -- Eine kurze Geschichte der KI -- Logisches Denken wird automatisch -- Systeme werden zu Experten -- Computer lernen sprechen -- Algorithmen simulieren die Evolution -- Neuronale Netze simulieren Gehirne -- Roboter werden sozial -- Automobile werden autonom -- Fabriken werden intelligent -- Von der natürlichen über die künstliche zur Superintelligenz? Jeder kennt sie. Smartphones, die mit uns sprechen, Armbanduhren, die unsere Gesundheitsdaten aufzeichnen, Arbeitsabläufe, die sich automatisch organisieren, Autos, Flugzeuge und Drohnen, die sich selber steuern, Verkehrs- und Energiesysteme mit autonomer Logistik oder Roboter, die ferne Planeten erkunden, sind technische Beispiele einer vernetzten Welt intelligenter Systeme. Machine Learning verändert unsere Zivilisation dramatisch. Wir verlassen uns immer mehr auf effiziente Algorithmen, weil die Komplexität unserer zivilisatorischen Infrastruktur sonst nicht zu bewältigen ist. Aber wie sicher sind KI-Algorithmen? Diese Herausforderung wird in der 2.Auflage aufgegriffen: Komplexe neuronale Netze werden mit riesigen Datenmengen (Big Data) gefüttert und trainiert. Die Anzahl der dazu notwenigen Parameter explodiert exponentiell. Niemand weiß genau, was sich in diesen „Black Boxes“ im Einzelnen abspielt. Im Machine Learning benötigen wir mehr Erklärung (explainability) und Zurechnung (accountability) von Ursachen und Wirkungen, um ethische und rechtliche Fragen der Verantwortung (z.B. beim autonomen Fahren oder in der Medizin) entscheiden zu können! Seit ihrer Entstehung ist die KI-Forschung mit großen Visionen über die Zukunft der Menschheit verbunden. Sie ist bereits eine Schlüsseltechnologie, die den globalen Wettstreit der Gesellschaftssysteme entscheiden wird. „Künstliche Intelligenz und Verantwortung“ ist eine weitere zentrale Ergänzung der 2. Auflage: Wie sollen wir unsere individuellen Freiheitsrechte in der KI-Welt sichern? Dieses Buch ist ein Plädoyer für Technikgestaltung: KI muss sich als Dienstleistung in der Gesellschaft bewähren
Keywords: Artificial intelligence, Engineering -
(2019): User-Adaptive Interaction in Social Robots: A Survey Focusing on Non-physical Interaction. In: International Journal of Social Robotics 11 (1), S. 185-205. DOI: 10.1007/s12369-018-0485-4
DOI: https://doi.org/10.1007/s12369-018-0485-4 Abstract: This work presents a survey on the usage of user-adaptive techniques for human interface with Social Robots, with focus on non-physical interaction. The work is based on an analysis of a number of recent scientific works, and aims to uncover existing scientific and technological gaps which can serve as basis for future research and development work. User-adaptive systems consist of autonomous agents that are able to use some manner of information on their user in order to adapt to them. Through their adaptive nature, these systems have been shown to be easier to accept by end-users, and to lead to improvements in a myriad of objective and subjective performance measurements. Thus, in the context of a growing domestic Social Robot industry, it becomes of key importance to study the scientific and technological frontiers of this field. In order to uncover potential lines of future research, we propose a taxonomy for the classification of works, which we use to analyse the works under survey, exposing the current scientific frontiers of the area. Aiming to establish the overall readiness of the field, we also analyse the maturity of the works under survey, exposing the current technological level of the techniques at hand and discussing a number of technological challenges. (PsycINFO Database Record (c) 2019 APA, all rights reserved)
-
(2019): Learning to Have an Ear for Face Super-Resolution. Online verfügbar unter http://arxiv.org/pdf/1909.12780v1
Abstract: We propose a novel method to perform extreme (16x) face super-resolution by exploiting audio. Super-resolution is the task of recovering a high-resolution image from a low-resolution one. When the resolution of the input image is too low (e.g., 8x8 pixels), the loss of information is so dire that the details of the original identity have been lost. However, when the low-resolution image is extracted from a video, the audio track is also available. Because the audio carries information about the face identity, we propose to exploit it in the face reconstruction process. Towards this goal, we propose a model and a training procedure to extract information about the identity of a person from her audio track and to combine it with the information extracted from the low-resolution input image, which relates more to pose and colors of the face. We demonstrate that the combination of these two inputs yields high-resolution images that better capture the correct identity of the face. In particular, we show that audio can assist in recovering attributes such as the gender and the identity, and thus improve the correctness of the image reconstruction process. Our procedure does not make use of human annotation and thus can be easily trained with existing video datasets. Moreover, we show that our model allows one to mix low-resolution images and audio from different videos and to generate realistic faces with semantically meaningful combinations.
-
(2019): Cognitive Science and Artificial Intelligence: Simulating the Human Mind and Its Complexity. In: Cognitive Computing and Systems. Online verfügbar unter https://www.researchgate.net/publication/336467083_Cognitive_Science_and_Artificial_Intelligence_Simulating_the_Human_Mind_and_Its_Complexity, zuletzt geprüft am 16.05.2020
Abstract: This study encompassed around the interdisciplinary study of cognitive science in the field of artificial intelligence. Past as well as current areas of research have been highlighted such that better understating of the topic can be ensured. Furthermore, some of the present-day applications of cognitive science artificial intelligence have been discussed as these can be considered as the foundation for further improvement. Prior to discussion about future scopes, real-time complexities have been revealed.
-
(2019): Social Robots and Seniors: A Comparative Study on the Influence of Dynamic Social Features on Human–Robot Interaction. In: International Journal of Social Robotics 11 (1), S. 5-24. DOI: 10.1007/s12369-018-0488-1
DOI: https://doi.org/10.1007/s12369-018-0488-1 Abstract: Due to the world’s aging demographics and declining caregiver-to-senior ratio, socially assistive robots are being designed to assist seniors with activities of daily living. Beyond functionality, the integration of such robots also depends on the quality of the interaction between the robot and senior, including the provision of social support. Our research investigates the effectiveness of robot social features on human–robot interaction (HRI) with seniors when providing assistance. Specifically, we investigate how the dynamic social features of a robot, such as facial expressions and gestures, affect the interaction experience of cognitively impaired seniors during an assistive activity. In this paper, we present a comparative HRI study of seniors living with mild cognitive impairments preparing a cup of tea with the assistance from three different embodiments with varying social features. In particular, the platforms considered were a human-like robot, a character-like robot, and a tablet display. Our results show that a human-like robot having an expressive face and arm gestures significantly increased levels of engagement, positive affect, and perceived social intelligence during the interaction when compared to the other two platforms. Furthermore, even though all platforms had the same activity assistance functionality, participants preferred the human-like robot, treated it like a companion, perceived it to be more useful, and were more inclined to keep it in social or private areas in their homes. No statistically significant differences were found between interactions with the character-like robot and the tablet. (PsycINFO Database Record (c) 2019 APA, all rights reserved)
-
(2019): How to stop computers from being biased. The bid to prevent algorithms producing racist, sexist or class-conscious decisions. In: Financial Times, 13.02.2019. Online verfügbar unter www.ft.com/content/12dcd0f4-2ec8-11e9-8744-e7016697f225, zuletzt geprüft am 23.10.2019
Abstract: One of the first major court cases over how algorithms affect people’s lives came in 2012, after a computer decided to slash Medicaid payments to around 4,000 disabled people in the US state of Idaho based on a database that was riddled with gaps and errors.
-
(2019) : BayMax: A Smart Healthcare System Provide Services to Millennials Using Machine Learning Technique: 2019 International Conference on Smart Structures and Systems (ICSSS): 2019 International Conference on Smart Structures and Systems (ICSSS): Chennai, India: IEEE, S. 1-5
DOI: https://doi.org/10.1109/ICSSS.2019.8882844 Abstract: Health is the utmost importance to any human being. To take care of oneself in this busty world we need some assistance. The world is revolving with smart technologies so we need a smart health care assistance that tracks one's activities, moods and suggest precautious actions when required. The primary objective of this product is to be time efficient along with being as an easily accesses able personal health assistant. To build a cost efficient personal assistance with good quality, monitoring of the messages from user is necessary. At present they are managed in an impromptu way. The paper will put forward the use of a Personal assistance chatbot to handle the messages from users and collectively provide the appropriate answers to it. A model has been proposed to enumerate precautions for the health related issues.
-
(2019): Pro-active positioning of a social robot intervening upon behavioral disturbances of persons with dementia in a smart nursing home. In: COGNITIVE SYSTEMS RESEARCH, S. 160-174. DOI: 10.1016/j.cogsys.2019.03.002
DOI: https://doi.org/10.1016/j.cogsys.2019.03.002 Abstract: Behavioral disturbances of persons with dementia residing in a nursing home impose a significant burden on other residents and on the care staff. A social robot can provide an adequate technological support tool for the caregivers by approaching a resident that exhibits a behavioral disturbance. In this paper, we focus on how to position the robot in the nursing home, taking into account the profile and location of the residents. We minimize the time between the detection of a behavioral disturbance and the robot having arrived near the resident and starting an interaction scenario. Our algorithm is evaluated using realistic data that was collected during 3 months in two Belgian nursing homes. (PsycINFO Database Record (c) 2019 APA, all rights reserved)
-
(2019): Towards ethical guidelines of using telepresence robots in residential care. In: International Journal of Social Robotics, S. 1-9. DOI: 10.1007/s12369-019-00529-8
DOI: https://doi.org/10.1007/s12369-019-00529-8 Abstract: Robotic telepresence is a potential technology to help alleviating the loneliness of elderly people. The impacts of long-term use of telepresence robots in residential care are not well known. We were interested in how using a telepresence robot influences the resident, family members and care workers at a facility, and what challenges and solutions there are for wider adoption of such robots in residential care. With a telepresence robot Double, we arranged a series of three trials in two separate residential care facilities: one 12-week trial in a private facility and two successive 6-week trials in a public facility. In each trial, we installed the telepresence robot in a room of a long-term care home resident for communicating with her/his family members. Based on the results, telepresence robots do increase presence and possibly engagement of family members in residential care, but privacy is a central concern. The mobility of a telepresence robot is hard to utilize in residential care, and to be able to do so, ethical consideration and guidelines are needed. We provide a draft of such ethical guidelines. (PsycINFO Database Record (c) 2019 APA, all rights reserved)
-
(2019) : Researching Im/politeness in Face-to-Face Interactions: On Disagreements in Polish Homes In: Ogiermann, Eva; Garcés-Conejos Blitvich, Pilar (Hg.): From speech acts to lay understandings of politeness: Multilingual and multicultural perspectives: Cambridge: Cambridge University Press, S. 146-174
-
Ogiermann, Eva; Garcés-Conejos Blitvich, Pilar (Hg.) (2019): From speech acts to lay understandings of politeness. Multilingual and multicultural perspectives. Cambridge: Cambridge University Press
DOI: https://doi.org/10.1017/9781108182119 Abstract: Taking an up-to-date and truly global approach, this volume presents a wide range of phenomena in politeness research, and discusses key developments in the field. Covering eight major world languages as well as several language varieties, a team of leading scholars provide a multilingual and multicultural perspective on various speech acts and emic conceptualisations of politeness, and a diachronic view of the field. Most significantly, the volume focuses on the latest trends in the field, such as metapragmatic approaches to im/politeness, politeness and globalization, politeness in computer mediated communication, and politeness and prosody, spanning a wide range of methodologies and types of data, including naturally occurring conversations, role plays, email messages, social media, online discussion forums, ethnographic interviews, focus groups, questionnaires, experiments and language corpora.
-
Oprea, Alina; Xu, Dongyan (Hg.) (2019): Proceedings 2019 Network and Distributed System Security Symposium. Network and Distributed System Security Symposium. San Diego, CA, February 24-27, 2019. Reston, VA: Internet Society
-
(2019): Human Understanding of Robot Motion: The Role of Velocity and Orientation. In: International Journal of Social Robotics 11 (1), S. 75-88. DOI: 10.1007/s12369-018-0493-4
DOI: https://doi.org/10.1007/s12369-018-0493-4 Abstract: A general problem in human–robot interaction is how to test the quality of single robot behavior, in order to develop robust and human-acceptable skills. The most typical approach are user tests with subjective measures (questionnaires). We propose a new experimental paradigm that combines subjective measures with an objective behavioral measure, namely viewing times of images viewed as self-paced slide show. We applied this paradigm to human-aware robot navigation. With three experiments, we studied the influence of two aspects of robot motion: velocity profiles and the robot’s orientation. A decreasing velocity profile influenced the predictability of the observed motion, and robot orientations diverting from the robot’s motion vector caused reduced perceived autonomy ratings. We conclude that the viewing time paradigm is a promising tool for studying human-aware robot behavior and that the design of human-aware robot navigation needs to consider both the velocity and the orientation of robots.
-
(2019): "Ungezogene Roboter": Welche Manieren erübrigen sich für Assistenzsysteme?. Forschungsgutachten für PoliTE (2019)
-
(2019): The paradox of serving: Can genuine servant leadership gain followers’ respect for the leader? Evidence from Germany and Lithuania. In: German Journal of Human Resource Management: Zeitschrift für Personalforschung 33 (2), S. 113-136. DOI: 10.1177/2397002218793840
DOI: https://doi.org/10.1177/2397002218793840 Abstract: A growing body of research shows that servant leadership is related to important organizational outcomes. Yet far less attention has been given to the mechanisms of servant leadership. This is particularly true for those dimensions of servant leadership that are inherently paradoxical and run counter to the common notion of leaders as strong influencers. In an effort to contribute to servant leadership research, this study investigated the relation between perceived genuine servant leadership behaviours (i.e. standing back, humility, authenticity) and respect for the leader, reflecting followers’ acceptance of their leader’s influence. Survey data were collected from 438 employees from various organizations in Germany and Lithuania and analysed via multi-group structural equation modelling. Results showed that the positive relation between perceived genuine servant leadership and respect for the leader was sequentially mediated by followers’ perceptions of alignment with ideal leader prototype and personal identification with the leader. Overall, this study elucidates the interplay between leadership and service, enhancing the understanding of how perceived genuine servant leadership has positive effects on followers. By drawing on two samples from different cultural clusters, this study strengthens the validity of the results and thus also contributes more generally to the cross-cultural applicability of servant leadership.
-
(2019): Angemessen ist, was als angemessen passiert. Forschungsgutachten für poliTE
-
(2019): Gemeinsame Welten. Zur Ontologie der menschlichen Wirklichkeit. Dissertation. Stuttgart
-
(2019) : Computational Tools for Human-Robot Interaction Design: 2019 14th ACM/IEEE International Conference on Human-Robot Interaction (HRI): Daegu, Korea: IEEE, S. 733-735
DOI: https://doi.org/10.1109/HRI.2019.8673221 Abstract: Robots must exercise socially appropriate behavior when interacting with humans. How can we assist interaction designers to embed socially appropriate and avoid socially inappropriate behavior within human-robot interactions? We propose a multi-faceted interaction-design approach that intersects human-robot interaction and formal methods to help us achieve this goal. At the lowest level, designers create interactions from scratch and receive feedback from formal verification, while higher levels involve automated synthesis and repair of designs. In this extended abstract, we discuss past, present, and future work within each level of our design approach.
Keywords: Angemessen(heit) (von Technik), automated synthesis, computational tools, Design methodology, Electric breakdown, Formal Methods, formal verification, human-robot interaction, human-robot interaction design, ieee xplore, Interaction Design, interaction designers, Maintenance engineering, multifaceted interaction-design approach, Programming, Robots, socially appropriate behavior, Task Analysis -
(2019): Machine behaviour. In: Nature 568 (7753), S. 477-486. DOI: 10.1038/s41586-019-1138-y
DOI: http://www.ncbi.nlm.nih.gov/pubmed/31019318 Abstract: Machines powered by artificial intelligence increasingly mediate our social, cultural, economic and political interactions. Understanding the behaviour of artificial intelligence systems is essential to our ability to control their actions, reap their benefits and minimize their harms. Here we argue that this necessitates a broad scientific research agenda to study machine behaviour that incorporates and expands upon the discipline of computer science and includes insights from across the sciences. We first outline a set of questions that are fundamental to this emerging field and then explore the technical, legal and institutional constraints on the study of machine behaviour.
-
(19.07.2019): Towards automatic estimation of conversation floors within F-formations, zuletzt geprüft am 01.10.2019
Abstract: The detection of free-standing conversing groups has received significant attention in recent years. In the absence of a formal definition, most studies operationalize the notion of a conversation group either through a spatial or a temporal lens. Spatially, the most commonly used representation is the F-formation, defined by social scientists as the configuration in which people arrange themselves to sustain an interaction. However, the use of this representation is often accompanied with the simplifying assumption that a single conversation occurs within an F-formation. Temporally, various categories have been used to organize conversational units; these include, among others, turn, topic, and floor. Some of these concepts are hard to define objectively by themselves. The present work constitutes an initial exploration into unifying these perspectives by primarily posing the question: can we use the observation of simultaneous speaker turns to infer whether multiple conversation floors exist within an F-formation? We motivate a metric for the existence of distinct conversation floors based on simultaneous speaker turns, and provide an analysis using this metric to characterize conversations across F-formations of varying cardinality. We contribute two key findings: firstly, at the average speaking turn duration of about two seconds for humans, there is evidence for the existence of multiple floors within an F-formation; and secondly, an increase in the cardinality of an F-formation correlates with a decrease in duration of simultaneous speaking turns.
-
(2019): From Impressions to Intentions: Direct and Indirect Effects of Police Contact on Willingness to Report Crimes to Law Enforcement. In: Journal of Research in Crime and Delinquency 56 (3), S. 412-450. DOI: 10.1177/0022427818817338
DOI: https://doi.org/10.1177/0022427818817338 Abstract: Objectives: Crime reporting intentions have been associated with ratings of police legitimacy and effectiveness. Less is known about the role of personal encounters with police. We explore this issue by specifying associations between reporting intentions and type of contact (involuntary/voluntary), scope (cumulative/recent), and appraisal (respect/satisfaction with last encounter). Methods: This study draws on surveys of young adults in New York City (N = 508) and uses structural equation modeling to examine correlates of crime reporting intentions. Results: Respondents with more stops are less willing to report, and this effect is both direct and indirect, operating primarily via legitimacy. The association between reporting intentions and variety of voluntary contacts is direct and positive. Negative appraisals of recent stops are associated with lower reporting intentions but only indirectly; feelings of disrespect in recent stops are mediated by perceptions of legitimacy. Dissatisfaction with a recent voluntary contact is related to negative views of police effectiveness, but effectiveness does not shape reporting intentions. Instead, disrespectful voluntary contact has a direct negative relationship with reporting attitudes. Conclusion: The results underscore the importance of better-specifying contact with the police, as reporting intentions and related perceptions of law enforcement vary across experience and appraisal mechanisms.
-
(2019): Regularity Extraction Across Species: Associative Learning Mechanisms Shared by Human and Non-Human Primates. In: Topics in cognitive science 11 (3), S. 573-586. DOI: 10.1111/tops.12343
DOI: https://doi.org/10.1111/tops.12343 Abstract: Extracting the regularities of our environment is a core cognitive ability in human and non-human primates. Comparative studies may provide information of strong heuristic value to constrain the elaboration of computational models of regularity learning. This study illustrates this point by testing human and non-human primates (Guinea baboons, Papio papio) with the same experimental paradigm, using a novel online learning measure. For local co-occurrence regularities, we found similar patterns of regularity extraction in baboons and humans. However, only humans extracted the more global sequence structure. It is proposed that only the first result that is common to both species should be used to constrain models of regularity learning. The second result indicates that the extraction of global regularities cannot be accounted for by mere associative learning mechanisms and suggests that humans probably benefit from their language recoding abilities for extracting these regularities. We propose to use a comparative approach to address a series of remaining theoretical questions, which will contribute to the development of a general theory of regularity learning.
-
(2019): Understanding Legal Meaning through Word Embeddings. In: SSRN Electronic Journal. DOI: 10.2139/ssrn.3455747
DOI: https://doi.org/10.2139/ssrn.3455747 Abstract: For judges and legal scholars, the quest for meaning and the identification of methods appropriate for understanding word meaning animates volumes upon volumes of debate. Likewise, social scientists interested in studying the law have increasingly recognized the variations in choices over words as important barometers for understanding the law. In this paper, I suggest recent advances in computational linguistics -- notably, the efficient estimation of distributed representations of word meanings, or word embeddings -- offer a potentially transformative avenue by which to assess the strategy, choice, and impact of judicial language. Utilizing a corpus of more than one million federal and state appellate court decisions, I estimate word embeddings for the more than 400,000 most common words found in legal opinions. In a series of simple illustrations, I demonstrate the value of this treatment of word meaning for new avenues of law and politics research.
-
(2019): Sharing the workplace with robots? New tool helps designers create safer socially oriented machines. In: Cyberpsychology, Behavior, and Social Networking 22 (5), S. 360-361. DOI: 10.1089/cyber.2019.29151.ceu
DOI: https://doi.org/10.1089/cyber.2019.29151.ceu Abstract: This article aims to to describe the characteristics of current cyberpsychology research in Europe. In particular, CyberEurope aims at describing the leading research groups and projects running on the other side of the Ocean. Human safety is a primary concern in human–robot interaction (HRI). When there is physical contact between humans and robots, dangerous collisions are likely. The safety map helps users to determine if the robot they are designing is capable of inflicting specific injuries during unexpected collisions. They can also pinpoint the most dangerous areas in the robot’s workspace and compare their robot with others in terms of safety characteristics. As a result, designers have clear information at their fingertips about the injuries most likely to occur during operation. This helps to guide the hardware design process, and also contributes to safe control and motion planning for the robot being designed. (PsycINFO Database Record (c) 2019 APA, all rights reserved)
-
Salichs, Miguel (2019): Social Robotics. 11th International Conference, ICSR 2019, Madrid, Spain, November 26-29, 2019 : proceedings. Cham, Switzerland: Springer (Lecture notes in computer science Lecture notes in artificial intelligence)